The effecT of Special- educaTion VoucherS on public School ...

C E N T E R F O R C I V I C I N N O V A T I O NA T T H E M A N H A T T A N I N S T I T U T E

C C i

Civ

iC R

epo

RtN

o. 5

2 A

pril

2008

Jay P. GreeneSenior Fellow,Manhattan Institute for Policy Research

Marcus A. WintersSenior Fellow,Manhattan Institute for Policy Research

The effecT of Special- educaTion VoucherS

on public School achieVemenT:

evidence from florida’s mcKay Scholarship program

Publ

ishe

d by

Man

hatt

an In

stitu

te

The Effect of Special-Education Vouchers on Public School Achievement: Evidence from Florida’s McKay Scholarship Program

exeCutive SummaRy

This paper evaluates the impact of exposure to a voucher program for disabled students in Florida on the academic performance of disabled students who remain in the public school system. The authors utilize student-level data on the universe of public school students in the state of Florida from 2000-01 through 2004-05 to study the effect of the largest school voucher program in the United States, the McKay Scholarship Program for Students with Disabilities (McKay), on achievement in math and reading by students who have been diagnosed as disabled and remain in the public school system.

This paper is the first empirical evaluation of the impact of exposure to a voucher program designed to allow students with disabilities to enroll in schools other than their local public schools on the achievement of disabled students who remain in their local public schools. Vouchers for disabled students are the fastest-growing type in the United States. Programs similar to McKay are currently operating in Ohio, Georgia, and Utah and have been recently considered by other states.

Highlights of the study include:

• Public school students with relatively mild disabilities made statistically significant test score improvements in both math and reading as more nearby private schools began participation in the McKay program. That is, contrary to the hypothesis that school choice harms students who remain in public schools, this study finds that students eligible for vouchers who remained in the public schools made greater academic improvements as their school choices increased.

• Disabled public school students’ largest gains as exposure to McKay increased were made by those diagnosed as having the mildest learning disabilities. The largest category of students enjoying the greatest gains, known as Specific Learning Disability, accounts for 61.2% of disabled students and 8.5% of all students in Florida.

• The academic proficiency of students diagnosed with relatively severe disabilities was neither helped nor harmed by increased exposure to the McKay program.

Civ

ic R

epor

t 52

April 2008

about the authoRS

JaY p. Greene is endowed chair and head of the Department of Education Reform at the University of Arkansas and a senior fellow at the Manhattan Institute. He conducts research and writes about topics such as school choice, high school graduation rates, accountability, and special education.

Dr. Greene’s research was cited four times in the Supreme Court’s opinions in the landmark Zelman v. Simmons-Harris case on school vouchers. His articles have appeared in policy journals such as The Public Interest, City Journal, and Education Next; in academic journals such as The Georgetown Public Policy Review, Education and Urban Society, and The British Journal of Political Science; as well as in newspapers such as the Wall Street Journal and the Washington Post. He is the author of Education Myths (Rowman & Littlefield, 2005).

Dr. Greene has been a professor of government at the University of Texas at Austin and the University of Houston. He received a B.A. in history from Tufts University in 1988 and a Ph.D. from the Department of Government at Harvard University in 1995. He lives with his wife and three children in Fayetteville, Arkansas.

marcuS a. WinTerS is a senior fellow at the Manhattan Institute. He has performed several studies on a variety

of education policy issues including high-stakes testing, performance-pay for teachers, and the effects of vouchers

on the public school system. His research has been published in the journals Education Finance and Policy, Teachers

College Record, and Education Next. His op-ed articles have appeared in numerous newspapers, including the Wall

Street Journal, Washington Post, and USA Today. He received his B.A. in political science from Ohio University in

2002, and an M.A. in economics from the University of Arkansas in 2006. He is currently a Ph.D. candidate in the

Department of Economics at the University of Arkansas and is expected to complete in the spring of 2008. He resides

in Manhattan.


1. introduction2. The mcKay Scholarship program 3. data

Table 1. Summary Characteristics of McKay Program

Table 2. Summary of Number of McKay Accepting Private Schools

Within Radius of Public Schools

Table 3. Percent of Students in Each Disability Category, 2000

4. method5. results

Table 4. Effect of McKay Competition on Student Math Test Scores

Table 5. Effect of McKay Competition on Student Reading Test Scores

Table 6. Overall Results — School With Average Competition

Within 5 Miles

6. conclusionreferencesendnotes

CONTENTS1

4

5

7

8

10

11

12

Civ

ic R

epor

t 52

April 2008


1. INTrOduCTION

School-choicepolicieshaveplayedanimportantroleintheeducationpolicydebateover the last twodecades.Cur-rently, 21 school-voucher programs in 14 states providetaxpayer-funded scholarships to attend a private school

(Enlow2008).Morethanhalfofallstatescurrentlyallowstudentstoenrollingovernment-fundedcharterschoolsthatoperateoutsideofmanyoftherulesandregulationsofthepublicschoolsystemanddonothavemandatedcatchmentzonesthatdeterminewhomayenrollinthem.

Overthelastfewyears,voucherprogramsforstudentswithdisabili-tieshavebeenamongthefastest-growingchoicepolicies.In1998,Florida’sMcKayScholarshipProgramforStudentswithDisabilities(McKay) became thefirst of its kind tooffer generous taxpayer-fundedvouchersthatstudentsidentifiedasdisabledcanusetoat-tendaprivateschoolorapublicschoolotherthantheirlocalone.Fromthetimeitwasfirstimplementedstatewide,in2000–01,tothe2006–07schoolyear,McKayhasgrownfromserving970studentsin100privateschoolstoserving18,273studentsin811privateschools,makingitthelargestschool-choiceprogramintheUnitedStates.

Other states have recently followed Florida’s lead by offeringvoucherprograms similar toMcKay fordisabled students. Spe-cial-educationvoucherprogramsarecurrentlyoperatinginOhio,

Jay P. Greene & Marcus A. Winters

1

the effeCt of SpeCial-eduCation

vouCheRS on publiC SChool aChievement:

evidenCe fRom floRida’S mCKay

SCholaRShip pRogRam

Civ

ic R

epor

t 52

April 2008

particular student’sproficiency ispartlydeterminedby the proficiency of his fellow students—perhapsbecausesuchstudentsprovidepositiverolemodels,stimulateclassroomdiscussion,andsoon.Thus,astheoverallqualityofstudentswithina localpublicschooldeteriorates,somighttheacademicprogressofanyofitsindividualstudents.

Itcouldbeargued,however,thatlosingstudentstoschool-choiceprogramswouldactuallyincreasetheabilityofpublicschoolstoelevatestudentachieve-ment.If,asmanyteachergroupsandpublicschooladvocatesclaimindebatesunrelatedtoschoolchoice,thetruecostofeducatingastudentisgreaterthantheresourcesthataschoolreceivestoeducatehimandiffundingislargelyallocatedonaper-capitabasis,thenlosingstudentstoschool-choiceprogramscouldmeanmoreresourcesavailableforthestudentswhoremain.Afurtherbenefitcouldbethataspublicschoolenroll-mentdropsastheresultofschool-choicepolicies,somightclasssize.

Suchresourceargumentsapplywithspecialforcetospecialeducation.Itisfrequentlyarguedthatspecial-educationprogramsareuniformlyunderfundedandthatthelargeincreaseinthepercentageofstudentswhoaredisabledhasbeenasubstantialburdenonlocalpublicschools.Ifthecostofeducatingdisabledstudentstrulyexceedsthefundingprovided,thenre-movingaportionofthesestudentsfromlocalpublicschoolswouldtendtoincreasetheresourcesavailableonaper-capitabasis.

Thesecondschoolofthoughtabouttheseso-calledsystemiceffectsholdsthatschool-choicepoliciesmightactually improve the performance of local publicschoolsevenas,andpreciselybecause,suchpoliciesreducetheirresources(Nechyba2003).Manyofthoseinfavorofschool-choicepoliciesarguethatthecur-rentsystem,inwhichstudentsareassignedtoschoolsonthebasisoftheiraddress,providesthoseschoolswithacaptiveclienteleitfeelslittlepressuretotrulyeducate.Thisweaknessisespeciallypronouncedwhenstudentslackthemeanstomovetoanotherschoolzoneorattendaprivatealternative.Special-educationstudents,inparticular,mightsufferunderthissystemifprivateschoolsarehesitanttoadmitthembecause

Utah,Georgia,andArizona.AccordingtotheAllianceforSchoolChoice,bills to implementorexpandaspecial-educationvoucherprogramhavealsopassedoneorbothhousesoftheNevada,Wisconsin,andVirginialegislatures.

Thesubstantialgrowthofschool-choicepoliciesintheUnitedStateshascreatedgreatdemandforresearchevaluatingtheimpactoftheseprograms.Thoughweare interested in the impactofchoiceprogramsonthosestudentswhousethem,itmaybemoreimpor-tantthatweunderstandtheeffectofthesepoliciesonthosestudentswhoremaininthepublicschoolsystem,largelybecausezonedpublic schoolswill continuetoeducatethevastmajorityofstudentsintheUnitedStatesfortheforeseeablefuture.

Thispaperisnotthefirsttoevaluatetheimpactofexposuretoschool-choicepoliciesontheacademicperformanceof studentswho remain in thepublicschoolsystem.However,itisthefirst(ofwhichweare aware) toprovide evidenceof the impact of aprogram aimed exclusively at disabled students ontheacademicachievementofthosedisabledstudentswhoremaininlocalpublicschools.

Therearetwogeneralschoolsofthoughtconcerningtheimpactofschool-choicepoliciesonalltypesofstudentswhoremaininthepublicschoolsystem.Thefirstschoolofthoughtisthatschool-choiceprogramsdecrease public school performance by drainingpublic schools of substantial financial and humanresources.Asstudentsleavelocalpublicschoolsforalternatives,theytakewiththemalargeportionofthefundingthattheschoolwouldhavereceivedfortheireducation.Such lossesare likely toharmstudents’academicproficiency.

Additionally, it is often assumed that the best andbrightestwithinanyparticularpoolofvoucher-eligiblestudentsaremorelikelytotakeadvantageofvoucherprograms. Several theoretical models suggest thatremovingthesestudentsfrompublicschoolswouldtendtostunttheacademicgrowthofstudentsremain-inginthepublicschoolsastheresultofaso-calleddilution of peer effects (Epple and Romano 1998,2002;Nechyba1999,2000;Caucutt2001).Thatis,a

2


3

theyaremoredifficulttoeducate.Underthistheory,school-choicepoliciescreateamarketforeducationalservices in which local public schools must begintocompetefortheirstudentsandtheresourcesthattheydrawuponbyofferinganeducationalproductof equalquality to the alternative that thevouchermakesavailable.

Several studies have evaluated the relationship be-tweenexposuretoschool-choicepoliciesandpublicschoolperformance.Utilizingslightlydifferentmeth-ods,GreeneandWinters(2004),Chakrabarti(2005),FiglioandRouse(2006),WestandPeterson(2006),andGreene(2001)eachfoundthatcompetitionfromavoucherprograminFlorida(adifferentprogramfromtheoneevaluatedinthispaper)ledtopublicschoolgainsonmathandreadingtests.Hoxby(2001)foundthat public schools improved their performance inresponsetocompetitionfromcharterschools.

Ina related literature,agrowingbodyofempiricalresearchmeasurestheimpactofgreaterexposuretoschoolingoptionsonpublicschoolstudents’academicoutcomes.Hoxby(2000),BayerandMcMillan(2005),andHanushekandRivkin(2003)findevidencethatgreater competition among public school districts,oftenreferredtoasTieboutchoice,leadstoimprovedpublic schoolperformance, thoughMcHugh(2003)findslessevidenceofthiseffect.Hoxby(1994)andDee (1998) find positive effects from unsubsidizedprivateschoolcompetition,whileSander(1999)andMcMillan(2004)failtofindsuchaneffect.

Unfortunately, to date there is no quantitative re-search evaluating the impact on the performanceofpublicschoolsofdisabledstudents’exposuretoschool-choiceprograms.Thispaperbeginstofillthisvoidintheliterature.

Focusingonaprogramdirectedatdisabledstudentsmaybeinterestingforavarietyofreasons.First,asdiscussedabove,suchpoliciesrepresentsubstantialgrowthingeneralinU.S.voucherprogramsoverthelastfewyears.

Focusingontheimpactofaspecial-educationvoucherprogramisalsoworthwhilebecauseafrequentcriticism

ofvoucherprogramshasbeenthatprivateschoolswillnotacceptstudentswithdisabilitiesbecausetheyaredifficulttoeducateand,ifenrolled,woulddecreasethe average ability level of their student body andthustheircompetitiveadvantage(EppleandRomano1998,2002;Nechyba1999,2000;Caucutt2001;CullenandRivkin2003).Thissuggeststhatnotonlyhighlyselectiveprivateschoolsmayrefrainfromacceptingdisabledstudents,butalsothemorenumerousurbanprivateschoolsthateducateseriouslydisadvantagedpopulations.Ifprivateschoolsareunwillingtoaccepteventhosedisabledstudentswhosetuitionispaid,theMcKayprogramshouldhavelittle,ifany,competitiveeffectontheperformanceofpublicschools.

AfinalfeatureoftheMcKayprogramthatmakesitofparticularinteresttostudyisthesizeofbothitseligibleandparticipatingpopulations.Animportantcriticismofpreviousschool-choiceresearchisthattheirfocusonsmallprogramsmaynothaveproducedfindingsthatwouldremainvalidastheseprogramsgrewinscale.Forexample,in2006–07,studentsinonly21Floridapublicschoolswereeligible to receiveavoucher from theoft-studiedOpportunityScholarshipProgram(GreeneandWinters2004;Chakrabarti2005;FiglioandRouse2006;WestandPeterson2006).Incontrast,in2005–06,about15percentofalldisabledpublicschoolstudentsinFloridawereeligibletoreceiveaMcKayvoucher.1

ThispaperutilizesadatasetprovidedbytheFloridaDepartmentofEducationtostudytheimpactofin-creasedexposuretotheMcKayprogramonthepro-ficiencyofdisabledstudentswhoremainedinlocalFloridapublicschools.ThisdatasetallowsustofollowtheperformanceofeachindividualstudentenrolledingradesthreethroughteninaFloridapublicschoolfromthe2000–01tothe2004–05schoolyears.Ourabilitytotracktheperformanceofindividualstudentsovertime(knownaspaneldata)substantiallyimprovestheaccuracyofourestimatesbyallowingustodirectlycontrolforunobservablefactorsthatplayimportantrolesinastudent’sacademicprogress.

Further,thisrichdatasetenablesustodisaggregatetheimpactofexposuretoMcKayonstudentswithvaryingdegreesofdisability.Aswewilldiscussbelow,theabil-itytoclassifystudentsonthebasisoftheirparticular

Civ

ic R

epor

t 52

April 2008

disabilityallowsustobettercontrolstatisticallyfortheimpactofstudents’disabilityontheiracademicprofi-ciency.ThereissomereasontoexpectthatstudentswithdifferentdegreesofdisabilityhavemoreorfewerprivateschooloptionsundertheMcKayprogram.Ifitisthecasethatstudentswithminordisabilitiesarebetterabletoacquireaseatinaprivateschool,thenweshouldexpectthattheMcKayprogramhadadif-ferential impacton theproficiencyof studentswhoremainintheregularpublicschoolsystem.

Thoughmoreresearch isnecessary, thefindings inthispapersuggestthatgreaterexposuretotheMcKayprogram increased the ability of public schools inFlorida to produce educational gains for disabledstudents.OurresultsalsosuggestthattheimpactofexposuretoMcKaywasgreatestamongstudentswithminordisabilities—inparticular,thoseinthecategorySpecificLearningDisability,whichisbyfarthelargestspecial-educationcategoryinFloridaandintheUnitedStates.Moreover,innosubgroupwithinspecialeduca-tionwasstudentproficiencyharmed,onaverage,byincreasedexposuretotheMcKayprogram.

However, the nature of the analysis here does nothelptodiscovertheexactreasonsforsuchimprove-ment.Thesefindingscouldbecausedbyreallocationsto remaining students of funds made available bythedeparturesofvoucher recipients;orbyschoolsrespondingtoadditionalcompetitionresultingfromtheadventofamarketforstudents;orbysomeotherfactornotyetdiscussedintheliterature.Future“insidetheblackbox”researchisnecessarytodistinguishtherootcausesoftheimprovementsinstudentproficiencyresultingfromexposuretotheMcKayprogram.None-theless,thefindingsofthispapertendtosupporttheuseofvoucherprogramsfordisabledstudents.

Theremainderofthispaperisdividedintofiveadditionalsections.Section2brieflydescribestheMcKayprogramevaluatedhereanditsgrowthoverthelastseveralyears.Section3providesadescriptionofourrichlongitudinaldataset.WediscusstheempiricalapproachofthepaperinSection4,andSection5reports theresultsofourestimations.Finally,Section6summarizesourfindingsandprovidesageneraldiscussionoftheirimplicationsforthediscussionofschool-choicepolicies.

4

2. ThE MCKAy SChOlArShIP PrOGrAM

The John M. McKay Scholarship Program forStudentswithDisabilities is a statewidepro-graminFloridadesignedtoprovidedisabled

studentswiththeresourcestoattendeitherapublicschooldifferentfromtheonethattheirplaceofresi-dencewouldotherwisedictateoraprivateschoolthatacceptedthem.McKayscholarshipsareavailabletoanyFloridapublicschoolstudentwho:1)hasbeenassignedanIndividualEducationPlan(IEP)—essen-tially a judicially enforceable contract between theschool system and every student diagnosedwith adisability that specifies the services that the schoolsystemistoprovide;and2)wasenrolledintheFloridapublicschoolsystemduringthepreviousyear.OnceastudentusesaMcKayvoucher,heremainseligiblefortheprogramforaslongasheremainsinprivateschooloruntilheturns22yearsofage.

Forprivateschoolstoparticipateintheprogram,theymustmeetsafetyrequirementsandemployteacherswithatleastabachelor’sdegree.Unlikemanyotherschool-voucher programs, McKay does not requireprivate schools toaccept the sum that thevoucherrepresentsasfulltuitionpayment.

Importantly for our purposes, the McKay programhas undergone dramatic growth since it was firstimplementedasasmallpilotinthe1999–2000schoolyear.Table1reportssomebasicstatisticsabouttheprogramforeachyearofitsexistence.Between2000and2001,whentheprogramwasadoptedstatewide,and2006and2007,thenumberofstudentsusingaMcKayscholarshipincreasedfrom970to18,273,mak-ingitthelargestschool-voucherprogramofanykindinthenation.2Theincreaseinthenumberofstudentsisinlargepartduetotheincrease,from100to811,inthenumberofprivateschoolsinthatperiodwillingtoacceptthevoucher.

McKayisdistinguishablefromothervoucherprogramsnotonlybythesizeofitseligiblepopulationbutalsobythegenerosityofitsgrants.Everyeligiblestudentisprovidedwithavoucherthatisequivalentinvaluetothesumthathisoriginalpublicschoolwouldspendonhimifhedidnotuseitorthetuitionchargedby


5

theacceptingprivateschool,whicheverissmaller.Thismeansthatstudentswithmoreseveredisabilitieshavethechanceofreceivingalargervoucheramountthanstudentswithmilderdisabilitiesbecausetheformeraremoreexpensivetoeducate.AccordingtotheFloridaDepartmentofEducation,in2006–07McKayscholar-shipsrangedinvaluefrom$5,039to$21,907,withanaverageof$7,206.3

3. dATA

Weutilizeinformationfromarichdatasetprovided by the Florida Department ofEducation.Thisdatasetcontainsstudent-

levelinformationfortheuniverseofpublicschoolstu-dentswhowereenrolledingradesthreethroughtenintheFloridapublicschoolsystemfromthe2000–01tothe2004–05schoolyear.Foreachstudent-year,thedatasetcomprisesdemographicinformationandthechild’sscoreonthemathandreadingversionsoftheFlorida Comprehensive Assessment Test (FCAT)—astandardizedtestadministeredtoallstudentsingradesthreethroughten.Ifthestudentisdisabled,thedatasetreportsthestudent’sdisabilityclassification.Impor-tantlyforestimationpurposes,thedatasetincludesanidentificationnumberforeachstudentthatallowsustofollowhisperformanceovertime.

Thedatasetincludesinformationononlythosestu-dentswhowereenrolledinpublicschoolduringthetimeperiodcovered.Becauseprivateschoolstudents,includingthosewhoutilizeMcKayscholarships,are

notrequiredtotakethestate’sstandardizedexams,wehavenoinformationontheirperformanceaftertheyleavepublicschoolandthusnobasisforevaluatingMcKay’simpactonthem.4

Consequently,ourexpansivedatasetcanbeusedtoevaluatetheimpactofpublicschoolexposuretotheMcKayprogramontheproficiencyofonlydisabledstudentswithinthosepublicschools.“Exposure”totheprogramrefers to theexistenceandpervasive-ness of schooling alternatives reasonably availableto students assigned to a particular public schoolonthebasisof theirhomeaddress.Thisdefinitionof exposure does not depend on students’ usingavoucher to leaveaparticularpublicschool foraprivate alternative—though the number of privatealternativesnearbyandthenumberofstudentsleav-ingpublicschoolswithaMcKayvoucherarealmostcertainlyrelated—andwedonotdirectlycontrolforsuchattrition.Instead,weareinterestedinidentifyingtothebestofourabilityhowmanyfeasibleoptionswere available to students enrolled in a particularpublicschool,sincealargernumberofalternativesprovidesstudentswithmoreofanopportunitytotakeadvantageofthevoucherprogram.

Weadoptthestrategyofsomepreviousstudiesofthesystemiceffectofschool-choicepoliciesandutilizethenumberofprivateschoolswillingtoacceptaMcKayvoucherwithinareasonablegeographicradiusofapublicschoolasthemeasureofaschool’sexposuretotheprogram.StudentscouldtheoreticallyutilizeaMcKayvoucheratanyschoolinthestatewillingtoenrollthem;butasapracticalmatter,geographycanlimitastudent’sabilitytoattendprivateschool.ItcanthereforelimittheexposuretoMcKaythatapublicschoolfaces.ItshouldcomeasnosurprisethatpublicschoolswithanumberofprivateschoolsnearbythatarewillingtoacceptMcKayvouchersaredispropor-tionatelylikelytolosetheirstudentstotheprogram,whilepublicschoolssurroundedbyfew,ifany,suchschoolsareessentiallyunaffectedbytheprogram.Wecanthususethenumberofprivatealternativeswithinaparticularradiusofapublicschoolasaproxyfor,ormeasure of, the exposure that that school facesfromtheprogram.

Voucher-using Students

Voucher-accepting private Schools

2006-07 18,273 811

2005-06 17,300 751

2004-05 15,910 708

2003-04 13,739 687

2002-03 9,130 518

2001-02 5,013 296

2000-01 970 100

1999-00 2 1

Table 1. Summary Characteristics of McKay Program

Civ

ic R

epor

t 52

April 2008

Foreachyear inourdataset,weusedgeographi-calsoftwaretolocateeverypublicschoolandeveryprivateschool thathadregisteredwith thestateaswillingtoacceptaMcKayvoucher.Wethencountedseparately,foreachoftheyearsinthedataset,thenumber of private schools within five miles of agivenpublicschoolacceptingMcKayvouchers,andthenthenumberofsuchschoolswithintenmilesofeachpublicschool.5Usingauniqueschoolidentifierin the student-level data set, we then determinedthenumberofprivatealternativesavailabletoeachstudent within these two distances from his localpublicschoolduringeachyearcoveredbyourdata.Table2reportsdescriptivestatisticsbyyearonthenumberofMcKay-acceptingprivate schoolswithinafive-andaten-mileradiusoflocalpublicschoolsthatstudentsinourdatasetwould,intheabsenceofalternatives,attend.

Animportantfeatureofourdatasetisthatitestablishesnotonlythefactofastudent’sdisabilitybutalsoitsnature.FederallawspecifiesthedisabilitydiagnosesmeritinganIEP.Table3liststhecategoriesauthorizedbytheFloridaDepartmentofEducationandreportsthepercentageofallstudentsaswellasalldisabledstudents in the statewho fellwithineachcategoryduring the 1999–2000 school year, the year before

the McKay program was adopted statewide. ThesecategoriesincludeblindnessanddeafnessandrangeinseverityfromtherelativelymildSpecificLearningDisability (SLD) to Traumatic Brain Injury (TBI), aseveredisability.InFloridaandaroundthenation,byfarthelargestcategoryisSLD,accountinginFloridafor61.2percentofdisabledstudentsand8.5percentofallstudents.

Onereasonthattheidentificationofastudent’spar-ticulardisabilityisusefulisthatitenablesustodoabetterjobofstatisticallycontrollingfortheseverityofeachstudent’sdisabilitywhendeterminingproficiencyasmeasuredbytestscores.Previousresearch,whichcontrolledonlyforwhetherastudenthadbeendiag-nosedasdisabled,treatsidenticallytheimpactofevery

Within 5 miles Within 10 miles

2001-02 3.4 9.5

2002-03 5.5 15.5

2003-04 7.1 20.3

2004-05 7.1 20.3

Table 2. Summary of Number of McKay Accepting Private Schools Within radius

of Public Schools

percent of all Students percent of disabled Students

IEP 13.9%

Specific Learning Disability 8.5% 61.2%

Speech 1.5% 10.6%

Emotional 1.4% 9.7%

Language 1.0% 7.3%

Emotional Mental 0.6% 4.3%

Other Health Impairment 0.2% 2.1%

Emotional Severe 0.2% 1.7%

Deaf-Hearing 0.1% 0.9%

Orothopedic 0.1% 0.8%

Autistic <0.1% 0.3%

Visual-Blindness <0.1% 0.3%

Traumatic Brain Injury <0.1% 0.1%

Table 3. Percent of Students in Each disability Category, 2000

6


kindofdisabilityonstudents’academicachievement.Thatis,notingonlywhetherastudentisidentifiedasdisabledwithoutdisaggregatingbytypeofdisabilityignoresthedisparateeffectsofdegreesofdisability,frommildtosevere.

Anotherreasonthattheabilitytoidentifyaparticulardisabledstudent’sdiagnosisisimportantisthatMcKayvoucherscouldhaveeffectsontheacademicperfor-mance of students remaining in the public schoolsystemaccordingtotheirvariousdisabilityclassifica-tions.Inparticular,ifstudentswithmilderdisabilities(SLD,languagedisability,etc.)aremorelikelytofindprivateschoolswillingtoenrollthemthanstudentswithmoreseveredisabilities(blindness,severemen-talretardation,traumaticbraininjury,etc.),thentheirrespectivepublicschoolswouldfacedifferentlevelsofexposuretoMcKay.

AmongtheMcKay-eligible,studentsdiagnosedwithSLD,themildestclassification,likelyhavethegreat-est access toprivate school alternatives. SomeSLDstudents are not much different from non-disabledstudents,andthusposefewereducationalchallengesforprivateschools.Indeed,thereissomereasontobelieve thatasubstantialportionofstudents in theSLDcategorymaynotbedisabledatall.Singeretal.(1989)findsubstantialvariationacross thestates inthefunctionalabilitiesofstudentsidentifiedashavingamilddisability,indicatingthatnotallstatesfollowthesamemethodsforidentifyingstudents.MacMillanandSiperstein(2001)suggestthatpublicschoolsuselowachievementaloneinclassifyingstudentsasSLD.Privateschoolsmaybeparticularlywillingtoacceptstudents who suffer from little or no neurologicalobstacletohighachievementlateron.

Incontrast,wemightexpect thatstudentswithpar-ticularlysevereorraredisabilitieshavefewerprivateschoolalternatives,eventhoughtheMcKayvouchergrantislargerforstudentswithmoreseveredisabilities.Servingstudentswithcertaindisabilitiescouldrequireanoriginalfixed-costinvestmentinfacilitiesthatfewprivateschoolsarewillingtomake,leavingthemoreseverelydisabledstudentswithfewereducationalop-tionsunderMcKaythanstudentswithmilderdisabilities.Itis,however,importanttonotethatpreviousresearch

(GreeneandForster2003)findsthatthedistributionofdisabilitiesintheMcKayprogramresemblesthedistri-butioninpublicschools.

Students’differentialaccesstoprivateschoolingforthereasonsofferedaboveshouldalsogivepublicschoolsdifferentlevelsofexposuretotheMcKayprogram.Bydistinguishingeachstudent’sparticulardisabilityclas-sification,ourdatasetallowsustodeterminewhethertheimpactontheacademicachievementofdisabledstudentsoftheadditionofaMcKay-acceptingprivateschoolneartheirpublicschoolvariesaccordingtothenatureoftheirdisabilities.

4. METhOd

We evaluate whether students in schoolswith greater exposure to McKay vouch-ers—measuredbythenumberofprivate

schoolswithinacertainradiusoftheirpublicschoolthatarewillingtoacceptthevouchers—madelargerorsmalleracademicgainsinmathandreadingthanstudents in schools that faced less exposure to theprogram.Ouranalysisalsomeasures the impactofdifferencesinexposuretoMcKayonstudentsineachdisabilitycategory.

WeusethepaneldatasetthatincludesinformationontheuniverseofpublicschoolstudentsinFlorida,whether or not they are disabled. The dependentvariableintheanalyses—thatis,theoutcomeweareevaluating—isthestudent’s testscoreonthestate’smandatedmathorreadingexam.

Weutilizethisexpansivedatasettoestimatestudentproficiencyinmathandreadingovertime.Inessence,through the use of a so-called fixed-effects regres-sionmodel,wecancontrolforfactorsthatinfluenceastudent’sacademicproficiencybutareunobservedbytheresearcher—forexample,parentalinvolvement,nutrition,andwealth.Theregressionalsostatisticallycontrols forobservable factors important tostudentproficiency,suchas theschooldistrict thechildat-tends, theyearof theobservation,gradelevel,andobservable studentdemographic characteristics thatvaryovertime,suchaswhetherthestudentiseligible

7

Civ

ic R

epor

t 52

April 2008

private Schools Within 5 miles private schools within 10 miles

coefficient robust Standard

error

t coefficient robust Standard

error

t

Specific Learning Disability * Within Radius 2.30 0.10 23.58 *** 1.13 0.04 26.97 ***

Speech * Within Radius 0.24 0.17 1.38 0.14 0.07 2.14 **

Emotional * Within Radius 1.20 0.24 4.97 *** 0.56 0.11 5.36 ***

Language * Within Radius 2.78 0.26 10.54 *** 1.19 0.11 11.28 ***

Emotional Mental * Within Radius 0.86 0.47 1.83 * 0.31 0.19 1.60

Other Health Impairment * Within Radius 0.97 0.35 2.73 *** 0.41 0.14 2.99 ***

Emotional Severe * Within Radius 0.87 0.48 1.83 * 0.41 0.19 2.14 **

Deaf-Hearing * Within Radius 1.98 0.75 2.64 *** 1.18 0.31 3.77 ***

Orthopedic * Within Radius 1.51 0.95 1.59 0.64 0.40 1.61

Autistic * Within Radius 0.75 1.12 0.67 0.37 0.42 0.88

Visual-Blindness * Within Radius -0.49 1.36 -0.36 0.14 0.50 0.28

Traumatic Brain Injury * Within Radius -0.41 2.28 -0.18 -0.51 0.97 -0.53

Within R-Square 0.5240 0.5242

Number of Observations 6,219,617 6,219,617

Number of Groups 2,398,331 2,398,331

forafreeorreduced-price lunchor is identifiedashavinglimitedEnglishproficiency.

Wedomorethanmeasuretheoveralleffectontheperformanceoftheaveragestudent(disabledornotdisabled)ofexposuretoMcKay:wealsomeasurethedifferentialimpactonstudentswithineachparticulardisabilityclassification.Wedosoby interacting thechild’sdisabilityclassificationwiththenumberofpri-vateschoolswithinagivenradiusofhispublicschool.ThesevariablesareanestimationoftheaverageeffectofincreasedexposuretotheMcKayprogramontheacademicproficiencyof students in eachparticulardisabilityclassification.6

WefirstestimatetheeffectbyusingasourexposurevariablethenumberofprivateschoolsacceptingMcKayvoucherswithinafive-mile radiusofagivenpublicschoolinaparticularyear.Inordertotestforthero-bustnessofourprocedure,wethenreplacethisnumberwiththenumberofprivateschoolsacceptingMcKayvoucherswithinaten-mileradiusofthesamepublic

school.Inordertoestablishthevalidityofafive-milecompass,welookfortheimpactofprivateschoolsthatarefartheraway.Wewouldexpectthattheeffectofincludingmoredistantprivateschoolswouldbesmallerthan the impactof thepresenceofmoreproximateprivateschoolsonpublicschoolperformance.

5. rESulTS

Theresultsofourestimationforstudentprofi-ciencyinmatharefoundinTable4.Aswillbethecasethroughout,forreasonsofspace

we report estimates for only the interaction of thediagnosisandcompetitionvariables.7

Our variables of interest are the interaction terms,whichevaluatethedifferentialeffectofMcKayexpo-sureonstudentswithdifferentdiagnoses.McKayex-posureandspecial-educationstudents’performanceinlightoftheirdiagnoses.Thevariablesaresortedinthetableaccordingtothesizeofthestudentpopulation

8

Table 4. Effect of McKay Competition on Student Math Test Scores

***Significant at p<.01 **Significant at p<.05 *Significant at p<.10


private Schools Within 5 miles private schools within 10 miles

coefficient robust Standard

error

t coefficient robust Standard

error

t

Specific Learning Disability * Within Radius 3.377997 0.1214082 27.82 *** 1.72 0.05 32.98 ***

Speech * Within Radius 0.0440374 0.2217133 0.2 0.04 0.09 0.47

Emotional * Within Radius 0.1922976 0.2775664 0.69 0.26 0.12 2.19 ***

Language * Within Radius 2.391393 0.3159964 7.57 *** 1.07 0.13 8.30 ***

Emotional Mental * Within Radius 1.585794 0.5015176 3.16 *** 0.67 0.21 3.24 ***

Other Health Impairment * Within Radius 1.270033 0.4155005 3.06 *** 0.47 0.16 2.86 ***

Emotional Severe * Within Radius 0.6513608 0.5621964 1.16 0.46 0.23 2.04 **

Deaf-Hearing * Within Radius 1.380605 0.866282 1.59 0.86 0.36 2.35 **

Orthopedic * Within Radius 0.7317186 1.038647 0.7 0.21 0.44 0.49

Autistic * Within Radius 2.655537 1.11179 2.39 ** 0.99 0.45 2.20 **

Visual-Blindness * Within Radius 1.747482 1.976992 0.88 0.71 0.79 0.89

Traumatic Brain Injury * Within Radius 0.7106963 2.323991 0.31 -0.59 1.01 -0.58

Within R-Square 0.3714 0.3716

Number of Observations 6,296,617 6,296,617

Number of Groups 2,406,010 2,406,010

inthespecial-educationcategory,whichisimperfectlyrelatedtotheseverityofthediagnosis.Theestimateoneachoftheinteractionvariablesispositive,andmostestimates (especially those for the less severecategories) are significantly different from (that is,almostcertainlylargerthan)zero.

Theresultsaresimilarforboththewithin-five-milesandthewithin-ten-milesspecifications.Theprimarydifferenceintheanalysesappearstobethattheesti-matesofthewithin-five-milesanalysisareuniformlylargerthantheestimatesofthewithin-ten-milesspeci-fication,whichisaswewouldexpectifstudentsareless likely to enroll inMcKay schools farther awayfromtheirhomesthanthepublicschoolinwhichtheywerepreviouslyenrolled.

Table5reportstheresultsofourestimationforread-ing.Theresultsaresimilartothosefoundformath,thoughtheimpactofMcKayexposureontheread-ing scoresofmost categoriesofdisabled students

appearstobelargerthanitwasonmathscores.Weagain see a positive relationship between McKayexposureandacademicproficiency inmostof thediagnosticcategories,especiallythemilderdisabili-ties.Asinmath,therelationshipuniformlydeclinesasweexpandtheexposurevariabletoincludeschoolswithintenmiles.

Table6putsourresultsofthewithin-five-mileses-timationintoamoremanageablecontext.ThetablefirstreportstheeffectofMcKayexposureonstudentswithineachclassificationinaschoolwiththeaver-age number of nearby private schools acceptingvouchers in 2004–05. The second column reportstheoveralleffectonthetestscoresofaschoolwithaverageMcKayexposurebyincorporatingtheesti-mateforboththeoveralleffectofMcKayexposureandtheinteractiontermforaparticulargroup.ThethirdcolumnstatestheoveralleffectonstudentsinaschoolwithaverageMcKayexposureasstandarddeviationunits.

9

Table 5. Effect of McKay Competition on Student reading Test Scores

***Significant at p<.01 **Significant at p<.05 *Significant at p<.10

Civ

ic R

epor

t 52

April 2008

differential effect from

average exposure

overall effect from

average exposure

overall effect from average exposure in

Standard deviation units

differential effect from

average exposure

overall effect from

average exposure

overall effect from average exposure in Standard deviation

units

Specific Learning Disability 16.31 15.99 0.05 *** 23.98 24.47 0.07 ***

Speech 1.68 1.36 0.00 0.31 0.80 0.00

Emotional 8.52 8.20 0.03 *** 1.37 1.85 0.01

Language 19.71 19.38 0.06 *** 16.98 17.46 0.05 ***

Emotional Mental 6.10 5.77 0.02 * 11.26 11.74 0.03 ***

Other Health Impairment 6.87 6.54 0.02 *** 9.02 9.50 0.03 ***

Emotional Severe 6.20 5.88 0.02 * 4.62 5.11 0.01

Deaf-Hearing 14.05 13.72 0.04 ** 9.80 10.29 0.03 *

Orthopedic 10.75 10.43 0.03 5.20 5.68 0.02

Autistic 5.33 5.01 0.02 18.85 19.34 0.05 **

Visual-Blindness -3.47 -3.79 -0.01 12.41 12.89 0.04

Traumatic Brain Injury -2.92 -3.24 -0.01 5.05 5.53 0.02

10

ThetableshowsthattheeffectonstudentsinaschoolwithaverageMcKayexposuredifferedbydisabilityclassification.Theeffectforstudentsidentifiedashav-ingaSpecificLearningDisability,whoconstituteabout8.5percentofallstudentsinFloridaand61percentofstudentsinspecialeducation(seeTable2),wasabouta0.05standarddeviationincreaseinmathanda0.07standarddeviationincreaseinreading.

6. CONCluSION

Thispaperaddstoanimportantandgrowingliterature evaluating the impact of school-choicepoliciesontheperformanceofpublic

schools.OurresultsfromevaluatingFlorida’sMcKayprogramprovideadditionalevidencethatratherthanbeingharmed,publicschoolsrespondtothechal-lengeofexposuretoschoolchoicebyimprovingthe

educationtheyprovide.Thesefindingsareconsistentwith most previous research, which demonstratesschool-choice policies’ positive effect on publicschoolachievement.

More specifically, this paper has provided the firstquantitative evaluation of the impact of a voucherprogram on disabled students. Such research is ofgrowingimportance,giventhesubstantialgrowthinschool-choiceprogramsaimedatthisparticulardis-advantagedpopulation.Theseinitialresultssupporttheuseofspecial-educationvoucherstoimprovetheeducationaloutcomesofdisabledstudentswithinpub-licschools.Muchmorequantitativeresearchontheseincreasinglyimportantpoliciesisnecessarytoprovideafullerunderstandingoftheireffectsbothonstudentswhoremaininpublicschoolsandonstudentswhousesuchvoucherstoattendprivateschools.

Table 6. Overall results — School With Average Competition Within 5 Miles math reading

Average 7.1 McKay Accepting Private Schools Within 5 Miles in 2004-05 Standard Deviation on FCAT Math test in 2004-05 = 311.2186 Standard Deviation on FCAT Reading test in 2004-05 = 389.9618

***Significant at p<.01 **Significant at p<.05 *Significant at p<.10Significance of overall relationship tested with F-test


11

math reading

RefeRenCeS

Bayer, P., & McMillan, R. (2005). “Choice and competition in local education markets.” Working paper 11802. NBER.

Caucutt, E. (2001). “Peer group effects in applied general equilibrium.” Economic Theory, 17, 25–51.

Chakrabarti, R. (2005). “Impact of voucher design on public school performance: Evidence from Florida and Milwaukee

voucher programs.” Working paper.

Cullen, J. B. (2003). “The impact of fiscal incentives on student disability rates.” Journal of Public Economics,

87(7), 1557–89.

———, & Rivkin, S. G. (2003). “The role of special education in school choice.” In “The economics of school choice,”

ed. Caroline Hoxby. NBER, 67–106.

Dee, T. (1998). “Competition and the quality of public schools.” Economics of Education Review, 17(4), 419–27.

Enlow, Robert C. (2008). “Grading school choice: Evaluating school choice programs by the Friedman gold standard.”

Friedman Foundation.

Epple, D., & Romano, R. (1998). “Competition between public and private schools, vouchers and peer group effects.”

American Economic Review, 88(1), 33–62.

——— (2002). “Educational vouchers and cream skimming.” Working paper. Carnegie Mellon University.

Figlio, D. N., & Rouse, C. E. (2006). “Do accountability and voucher threats improve low-performing schools?” Journal

of Public Economics, 90(1), 239–55.

Greene, J. P., (2001). “The looming shadow.” Education Next, 1(4), 76-82.

Greene, J. P. & Forster, G. (2003). “Vouchers for special education students: An evaluation of Florida’s McKay

scholarship program” Manhattan Institute, Civic Report No. 38.

Greene, J. P., & Winters, M. A. (2004). “Competition passes the test.” Education Next, 4(3), 66–71.

Hanushek, E., & Rivkin, S. (2003). “Does public school competition affect teacher quality?” and “Introducing school

choice into multidistrict public school systems.” In “The economics of school choice,” ed. Caroline Hoxby. NBER, 23–48.

Hoxby, C. (1994). “Do private schools provide competition for public schools?” Working paper 4978. NBER.

——— (2000). “Does competition among public schools benefit students and taxpayers?” American Economic

Review, 90(5), 1209–38.

——— (2001). “Rising Tide.” Education Next, 1(4), 69–74.

MacMillan, D. L., & Siperstein, G. N. (2001). “Learning disabilities as operationally defined by schools. Paper presented

at the 2001 LD Summit: Building a Foundation for the Future. Washington, D.C.

McHugh, C. (2003). “Does competition among public schools benefit students and taxpayers?: Comment.” Mimeo.

Duke University.

McMillan, R. (2004). “Competition, incentives and public school productivity.” Journal of Public Economics, 88(9), 1871–92.

Civ

ic R

epor

t 52

April 2008

12

Nechyba, T. (1999). “School finance induced migration patterns: The impact of private school vouchers.” Journal of

Public Economic Theory, 1(1), 5–50.

——— (2000). “Mobility, targeting and private school vouchers.” American Economic Review, 90(1), 130–46.

——— (2003). “Introducing school choice into multidistrict public school systems.” In “The economics of school

choice,” ed. Caroline Hoxby. NBER, 145–94.

Rouse, C. E., Hannaway, J., Goldhaber, D., & Figlio, D. (2007). “Feeling the Florida Heat? How Low-Performing

Schools Respond to Voucher and Accountability Pressure.” National Center for Analysis of Longitudinal Data in

Education Research, Working Paper 13.

Sander, W. (1999). “Private schools and public school achievement.” Journal of Human Resources, 34(4), 697–709.

Singer, J. D., Palfrey, J. S., Butler, J. A., & Walker, D. K. (1989). “Variation in special education classification across

school districts: How does where you live affect what you are labeled?” American Educational Research Journal,

26, 261–81.

West, M. R., & Peterson, P. E. (2006). “The efficacy of choice threats within school accountability systems: Results

from legislatively induced experiments.” The Economic Journal, 116, C46–C62.

endnoteS

1. Digest of Education Statistics 2006, table 50.

2. http://www.floridaschoolchoice.org/Information/McKay/files/Fast_Facts_McKay.pdf.

3. Ibid.

4. Many private school students are administered standardized exams in accordance with their school’s policy.

However, this information is not reported to the state.

5. Private schools that do not accept McKay vouchers are not counted as providing additional exposure to a public

school. An increase in the number of private alternatives faced by a public school from one year to the next is most

often the result of an existing private school’s decision to receive McKay vouchers and not of the opening of a brand-

new private school in the area.

6. We have identified three potential biases that might distort estimation of this regression: nonrandom attrition

of students who utilize McKay vouchers; nonrandom selection of students into special-education programs due to

changes in the financial incentives brought about by McKay; and nonrandom decisions of private schools to accept

McKay vouchers. These issues are discussed in detail in the more technical version of the paper found online at: www.

manhattan-institute.org/html/cr_52.htm. In each of these cases, we argue, and provide evidence, that these biases will tend

to underestimate the true impact of McKay exposure on student academic achievement.

7. Full results are available from the authors by request.

The Center for Civic Innovation’s (CCI) mandate is to improve the quality of life in cities by

shaping public policy and enriching public discourse on urban issues. The Center sponsors

studies and conferences on issues including education reform, welfare reform, crime reduction,

fiscal responsibility, Immigration, counter-terrorism policy, housing and development, and

prisoner reentry. CCI believes that good government alone cannot guarantee civic health,

and that cities thrive only when power and responsibility devolve to the people closest to

any problem, whether they are concerned parents, community leaders, or local police.

www.manhattan-institute.org/cci

The Manhattan Institute is a 501(C)(3) nonprofit organization. Contributions are tax-

deductible to the fullest extent of the law. EIN #13-2912529

CenteR foR CiviC innovation

Stephen Goldsmith, Advisory Board Chairman Emeritus

Howard Husock, Vice President, Policy Research

Erin A. Crotty, Associate Director

fellowS

Edward Glaeser Jay P. Greene

George L. Kelling Edmund J. McMahon

Peter Salins Fred Siegel

Marcus A. Winters

Date post:	31-Jan-2022
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

The effecT of Special- educaTion VoucherS on public School ...

Documents